This project involves methodological research on the analysis of 2x2 contingency table data. The data could originate from a case-control study, or as the result of a vaccine trial. Methods to estimate and test the odds ratio for case-control data or the vaccine efficacy will be investigated. One project constructed unbiased confidence intervals for the odds ratio of case-control exposure data. A method based on the profile likelihood was found to give shorter intervals than existing methods.